In December 2010 UBS came out with XVIX—an ETN that holds a long position in medium term volatility (equivalent to VXZ), and short a 50% position in short term volatility (equivalent to VXX). See Volatilty Tickers for the full set of available ETNs. XVIX&8217;s strategy was to take advantage of the historical situation (2004 to 2009) where short term futures were hugely eroded by contango and mid term futures could be used to protect the position from volatility spike. The product backtested great, showing returns of 24% and 55% percent in 2009 and 2010 respectively.
Since its introduction at $25 XVIX has gone no where, just some drifting around, mostly down.
The primary reason for XVIX&8217;s poor performance is the term structure change in medium term volatility futures that occurred in 2010. In the graph below, negative percentages indicate that the futures are in contango, positive numbers indicate backwardation.
Even though VIX in the Summer of 2011 dropped below 15 and VXX dropped almost to 20 the medium term futures stayed in contango. The next chart adds the short term contango over the same time frame scaled so that the maximiums / minimums are about the same. The contango / backwardation on the short term 1 / 2 month futures used to be about 5X larger than the medium term equivalent—now it only about 2X larger.
While the behavior of short term contango has stayed about since the beginning of the bull market in March 2009 the contango of the medium term futures has tended to increase, with the average value in 2011 being more negative than 2010. This looks like a structural shift in the term structure and it isn&8217;t good news for XVIX. Unless UBS amends its mix of medium and short term futures, it looks like XVIX will be a long term loser.
If you are interested in downloading the data used for this analysis, see this post.
Saturday, December 1st, 2012 |